| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Thu, 23 Nov 2023 19:10:17 +0200
From: claudiu beznea <claudiu.beznea@...on.dev>
To: Geert Uytterhoeven <geert@...ux-m68k.org>
Cc: s.shtylyov@....ru, davem@...emloft.net, edumazet@...gle.com,
kuba@...nel.org, pabeni@...hat.com, p.zabel@...gutronix.de,
yoshihiro.shimoda.uh@...esas.com, wsa+renesas@...g-engineering.com,
biju.das.jz@...renesas.com,
prabhakar.mahadev-lad.rj@...renesas.com,
sergei.shtylyov@...entembedded.com,
mitsuhiro.kimura.kc@...esas.com, masaru.nagai.vx@...esas.com,
netdev@...r.kernel.org, linux-renesas-soc@...r.kernel.org,
linux-kernel@...r.kernel.org,
Claudiu Beznea <claudiu.beznea.uj@...renesas.com>
Subject: Re: [PATCH 08/13] net: ravb: Rely on PM domain to enable refclk
Hi, Geert,
On 23.11.2023 10:48, Geert Uytterhoeven wrote:
> Hi Claudiu,
>
> Thanks for your patch (which seems to have been delayed by 3 days, ouch)!
>
> On Thu, Nov 23, 2023 at 5:35 AM Claudiu <claudiu.beznea@...on.dev> wrote:
>> From: Claudiu Beznea <claudiu.beznea.uj@...renesas.com>
>>
>> For RZ/G3S and RZ/G2L SoCs the Ethernet's reference clock is part of the
>> Ethernet's power domain. It is controlled though CPG driver that is
>> providing the support for power domain that Ethernet belongs. Thus,
>> to be able to implement runtime PM (at least for RZ/G3S at the moment)
>
> Why only for RZ/G3S?
(I'm copy pasting here what I already replied to Sergey)
The reasons I've limited only to RZ/G3S are:
1/ I don't have all the platforms to test it
2/ on G1H this doesn't work. I tried to debugged it but I don't have a
platform at hand, only remotely, and is hardly to debug once the
ethernet fails to work: probe is working(), open is executed, PHY is
initialized and then TX/RX is not working... don't know why ATM.
>
>> w/o the need to add clock enable/disable specific calls in runtime PM
>> ops of ravb driver and interfere with other IP specific implementations,
>> add a new variable to struct_hw_info and enable the reference clock
>> based on the value of this variable (the variable states if reference
>> clock is part of the Ethernet's power domain).
>>
>> Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@...renesas.com>
>
>> --- a/drivers/net/ethernet/renesas/ravb.h
>> +++ b/drivers/net/ethernet/renesas/ravb.h
>> @@ -1043,6 +1043,7 @@ struct ravb_hw_info {
>> unsigned nc_queues:1; /* AVB-DMAC has RX and TX NC queues */
>> unsigned magic_pkt:1; /* E-MAC supports magic packet detection */
>> unsigned half_duplex:1; /* E-MAC supports half duplex mode */
>> + unsigned refclk_in_pd:1; /* Reference clock is part of a power domain. */
>> };
>>
>> struct ravb_private {
>> diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
>> index 836fdb4b3bfd..ddd8cd2c0f89 100644
>> --- a/drivers/net/ethernet/renesas/ravb_main.c
>> +++ b/drivers/net/ethernet/renesas/ravb_main.c
>> @@ -2502,6 +2502,7 @@ static const struct ravb_hw_info gbeth_hw_info = {
>> .tx_counters = 1,
>> .carrier_counters = 1,
>> .half_duplex = 1,
>> + .refclk_in_pd = 1,
>> };
>>
>> static const struct of_device_id ravb_match_table[] = {
>> @@ -2749,12 +2750,14 @@ static int ravb_probe(struct platform_device *pdev)
>> goto out_release;
>> }
>>
>> - priv->refclk = devm_clk_get_optional(&pdev->dev, "refclk");
>> - if (IS_ERR(priv->refclk)) {
>> - error = PTR_ERR(priv->refclk);
>> - goto out_release;
>> + if (!info->refclk_in_pd) {
>> + priv->refclk = devm_clk_get_optional(&pdev->dev, "refclk");
>> + if (IS_ERR(priv->refclk)) {
>> + error = PTR_ERR(priv->refclk);
>> + goto out_release;
>> + }
>> + clk_prepare_enable(priv->refclk);
>> }
>> - clk_prepare_enable(priv->refclk);
>
> Is this patch really needed? It doesn't hurt to manually enable a
> clock that is also under Runtime PM control. Clock prepare/enable
> refcounting will take care of that.
I agree with that. I chose this path to not interfere w/ the comments
ravb_runtime_nop() which I didn't understand. Also I fail to understand why
the ravb_runtime_nop() is there...
>
>>
>> if (info->gptp_ref_clk) {
>> priv->gptp_clk = devm_clk_get(&pdev->dev, "gptp");
>> @@ -2869,7 +2872,8 @@ static int ravb_probe(struct platform_device *pdev)
>> if (info->ccc_gac)
>> ravb_ptp_stop(ndev);
>> out_disable_refclk:
>> - clk_disable_unprepare(priv->refclk);
>> + if (!info->refclk_in_pd)
>> + clk_disable_unprepare(priv->refclk);
>> out_release:
>> free_netdev(ndev);
>> pm_runtime_put:
>> @@ -2890,7 +2894,8 @@ static void ravb_remove(struct platform_device *pdev)
>> if (info->ccc_gac)
>> ravb_ptp_stop(ndev);
>>
>> - clk_disable_unprepare(priv->refclk);
>> + if (!info->refclk_in_pd)
>> + clk_disable_unprepare(priv->refclk);
>>
>> /* Set reset mode */
>> ravb_write(ndev, CCC_OPC_RESET, CCC);
>
> Gr{oetje,eeting}s,
>
> Geert
>
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@...ux-m68k.org
>
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
> -- Linus Torvalds
Powered by blists - more mailing lists